NATURE OF THE GENETIC EFFECTS 431 



of course be violated for a time, but it cannot be \dolated indefinitely, in 

 the direction of increase, without finally causing extinction of the popula- 

 tion. When, for example, individuals who would otherwise have been 

 eliminated are saved for reproduction by medical and other artificial aids, 

 the elimination rate temporarily falls below that of origination of new 

 mutant genes, so that the frequency of mutant genes in the population is 

 gradually raised. This increase continues until individuals become 

 heavily enough afflicted to undergo genetic elimination, despite the 

 artificial aids, at a rate which is again equivalent to the old rate of 2^, 

 modified by the overlapping, at which they are still receiving new 

 mutant genes. Hence 0.2 or some related value, most likely higher, 

 remains the figure representing the necessary long-term genetic elimina- 

 tion rate in man as long as the mutation rate remains w^hat it now is. 



The figure for elimination rate represents also the average amount of 

 genetic disability suffered in the long run by the individuals of the 

 population. That is, utilizing the above figure, the individuals must on 

 the average have an amount of disability commensurate with this one 

 chance in five of genetic extinction. The figure 0.2 does not mean that 

 one indi\'idual in five carries some one gene which regularly causes death 

 or failure to reproduce. For it was shown above that the great majority 

 of the genes which arise by mutation give, individually, only a small 

 chance of extinction but have a correspondingly high persistence, thereby 

 becoming distributed among many individuals. Thus each individual 

 comes to have many of these genes, enough to make the average indi- 

 vidual's chance of extinction, caused by all his mutant genes collectively, 

 one in five. The average frequency of the mutant genes per individual 

 of the population must (ignoring the occurrence of homozygotes) be 

 2/i times the average persistence p of these genes. Although we are far 

 from knowing the value of p, it is almost certainly several score and more 

 likely at least 100. That is, each mutant gene is probably distributed to 

 at least 100 individuals, on the average, before it dies out. With fx taken 

 as 0.3 this would cause each individual to carry, on the average, at least 

 60 deleterious heterozygous genes, usually of individually small effect.* 



In view of the fact that the distribution of the mutant genes with 

 respect to each other is on the whole a random one, the number of genes in 

 different individuals would tend to follow a Poisson (random) distribution. 

 Thus, with an average value of 60 per individual, the numbers in different 

 individuals would not range very widely about this, having a standard 



* It is true that the comparatively low figure of 8 was presented by the writer in an 

 earlier paper (Aluller, 1950b), but, as was there pointed out, this represented an 

 attempt to find a "rock-bottom" minimum value, based on assumptions that were 

 almost certainly too cautious. The above figure of 60, on the other hand, results 

 from an attempt to find a value based on assumptions regarding mutation rate and 

 gene action which seem more probable. 



